Sprint2 Library:

Team SELECT STARS: Alex Howard, Patrick Yang, Jade Yun, Tim Lee

Architecture

"""
# -------------#
# -------------#
# ----LAPTOP---#
# -------------#   ----> deploy.py -->  # -------------#
# -------------#                        # -------------#
        |                               # -----AWS-----#
       deploy.py                        # --WEBSERVER--#
        |                               # -------------#
# -------------#  ---> JSON POSTS ->    # -------------#
# -------------#                        # -------------#
# -TEST SERVER-#                            - Runs server.py
# -------------#                            - processes valid json
# -------------#                            - logs all requests RAW
												    - records locally in PROC
"""

How to run:

Here is an example of how to deploy the server

deploy('path_to_ssh_private_key.pem', 'server-address','prefix')

connected to ec2-35-166-134-236.us-west-2.compute.amazonaws.com
Already up-to-date.
repository repulled
web server started

How to stop:

Since gunicorn is running in the background, here's an easy way to terminate the server remotely.

python stop_aws_server.py server-address

	>checking for server ec2-35-166-134-236.us-west-2.compute.amazonaws.com
	>server already running, stopping

Files

file/dir	desc
archive_old_versions/	Previous prototype versions on different platforms
phase1/	All of sprint1's work, simple remote json processing
sample_json_generator/	scripts to generate synthetic json input data
deploy.py	deploy the json_catcher to aws
json_processing.py	utilities to validate incoming json and extract data
server.py	JSON ingestion web API built on flask
stop_remote_aws_server.py	stop the json_catcher remotely

Deployment Specifications:

• Remotely deploys this repo to a AWS instance
• Ensure that the ports are configured correctly with security
• Python: starts a Flask application that accepts HTTP post requests

API Webservice Specifications:

• Accepts raw data within the body (POST), expect formatted jsons
• Request Log (successful or not) /srv/runme/prefix/Raw.txt
• Use a logrotate or TimedRotatingFileHandler to rotate the file every 2 minutes into another file within the same directory
• If request is valid, then process into /srv/runme/prefix/proc.txt
• prefix will be passed to deploy function which will define the name of your output file

Testing service:

• Server #2 will test server #1 by sending repeated JSON posts
• Test Send Server: generate and send good and bad jsons

Reference PDF

Library Options:

• Flask
• Tornado (also python)

Project Planning, key tasks:

Task	Assignment
retrofit deploy	Tim
write the webpaths for requests	Alex
logging of web requestes	Patrick
create test server (to send)	Jade

Notes:

• Helpful to check web API and apps: https://www.getpostman.com/